Developing and testing a new dynamic assessment of late talkers' word-learning

Abstract

The current project brings together three bodies of literature: factors influencing infant and toddler word-learning, late talking in 2-year-old children, and dynamic assessment of language. These different bodies of research were integrated and used to support the development and testing of a new task with both theoretical and clinical applicability, the Dynamic Assessment of Late Talkers’ Word-Learning (DALT). A pilot study involving ten children was used to make further refinements and to determine readiness for progression to the main study. The main study adopted a quasi-experimental design. The DALT was administered as part of a battery of assessment to fifty-one children (19 late talking (LT), 8 within the borderline (BL) range, and 24 typically developing (TD)) when each was aged 24-months.

The DALT took the form of a scripted play activity within which four novel words were taught using three standardised levels of teaching support (cues). Children were awarded mediation scores based on the earliest point at which they were able to retain, identify a second or third exemplar of (i.e., transfer their understanding of), and produce each word. Children were awarded modifiability scores based on experimenter ratings of (child) responsivity and (experimenter) effort in administering the task.

Both mediation and modifiability scores were strongly correlated with standardised language assessment scores and significantly differentiated LT and TD children. When LT and BL children were grouped together to form a new ‘at risk’ group, classification accuracy for ‘at risk’ versus TD was 82.4% for mediation scores alone. The DALT reliably distinguished weaker and stronger word learners and showed potential as a useful screening and monitoring tool for LT children.

Findings added to current theories of the relationship between LT and TD children. Overlap between the mediation scores of higher scoring LT children and lower scoring TD children provided support for a dimensional account wherein LTs sit at the lower end of a LT-TD continuum. However, a sub-group of very low scoring LTs also provided support for a distinct category account wherein LT-TD differences are qualitative. Findings also added to the DA in speech and language therapy (SLT) literature by addressing barriers pertaining to clinical fidelity, adoption, and acceptability.

Date of Award	Feb 2023
Original language	English
Supervisor	Juliana Gerard (Supervisor), Catrin Rhys (Supervisor) & Lynda Kennedy (Supervisor)